mets (version 1.3.7)

resmeanATE: Average Treatment effect for Restricted Mean for censored competing risks data using IPCW

Description

Under the standard causal assumptions we can estimate the average treatment effect E(Y(1) - Y(0)). We need Consistency, ignorability ( Y(1), Y(0) indep A given X), and positivity.

Usage

resmeanATE(formula, data, model = "exp", outcome = c("rmst", "rmtl"), ...)

Arguments

formula

formula with 'Event' outcome

data

data-frame

model

exp ("exp") or identity link ("lin")

outcome

restricted mean time (rmst) or restricted mean time lost (rmtl)

...

Additional arguments to pass to binregATE

Author

Thomas Scheike

Details

The first covariate in the specification of the competing risks regression model must be the treatment effect that is a factor. If the factor has more than two levels then it uses the mlogit for propensity score modelling. We consider the outcome mint(T;tau) or I(epsion==cause1)(t- min(T;t)) that gives years lost due to cause "cause" depending on the number of causes. The default model is the exp(X^ beta) and otherwise a linear model is used.

Estimates the ATE using the the standard binary double robust estimating equations that are IPCW censoring adjusted.

Examples

library(mets); data(bmt); bmt$event <- bmt$cause!=0; dfactor(bmt) <- tcell~tcell
out <- resmeanATE(Event(time,event)~tcell+platelet,data=bmt,time=40,treat.model=tcell~platelet)
summary(out)

out1 <- resmeanATE(Event(time,cause)~tcell+platelet,data=bmt,cause=1,time=40,
                   treat.model=tcell~platelet)
summary(out1)

ratioATE(out,out1,h=function(x) log(x))